Free tool, no account required

Free GEO/AEO Page Evidence Checker

Inspect observable page signals that support retrieval, answer extraction, attribution and topic structure. Every result maps to evidence from the submitted page.

Also known as: GEO, AEO, AI visibility, generative engine optimization

Evidence shown with sourceNo account requiredNo invented metrics

Run a real check above

Submit a public URL, domain, or keyword. Novaverb will show only the evidence this tool can actually retrieve or measure.

Evidence model

Know what the result proves

This check reports observable page signals related to retrieval, answer extraction, attribution and topic structure. It counts passed and measured checks directly instead of converting them into a visibility score. It does not query an answer engine, observe a citation or prove that any system will use the page.

1. Source

Live page fetch with retrieve, quote, cite and graph analysis. The result identifies where its evidence came from.

2. Boundary

This reads the page's initial HTML plus /llms.txt and /robots.txt. It reports GEO/AEO page-level checks; it does not query AI engines or prove a specific citation.

3. Next action

Use the finding to verify a problem, then connect a workspace when you need history, monitoring, or site-wide analysis.

GEO explained

A practical GEO checker powered by observable evidence

Generative Engine Optimization (GEO) is the work of making a page easy for answer engines to retrieve, understand, quote and attribute. NovaBrain models that journey as four decisions: Retrieve, Quote, Cite and Graph.

How NovaBrain extracts the report

Retrieve - checks the response, HTML, indexability, canonical, robots.txt, sitemap and llms.txt.
Quote - measures readable words, headings, question framing, opening answer passages, lists, tables and metadata.
Cite - looks for JSON-LD, answer schemas, author/date signals, authority links, social metadata and transport trust.
Graph - maps internal and external links, entity schema, breadcrumbs, language and alternate-language hints.

What the checks mean

The checks describe observable page facts, not a ranking or citation promise. They show which evidence is present, what needs review and where a clear gap exists. The full Novaverb workspace can extend this single-page preview into a crawl, compare changes over time, connect evidence to tasks and track citation outcomes only when that evidence is available.

GEO, SEO and AEO

SEO focuses on search visibility, AEO on direct answers, and GEO on generated answers that combine and cite sources. The same fundamentals support all three: clear answers, durable structure, accurate schema, trustworthy sources and useful internal links.

This checker reads the page's HTML plus its /llms.txt and /robots.txt to report retrieve, quote, cite and graph checks. It does not query AI engines or prove a specific citation.

Get the right answer

What to submit - and what to avoid

Submit one page address, because retrievability is a per-page property. Scheme, www, path and query are normalised. A page that renders its content and its structured data in JavaScript may be reported as unready, which is also how an engine that does not execute scripts sees it.

Use it like this

yourdomain.com/guideAny address form works - a content page you want AI engines to retrieve and cite. http or https, with or without www, a bare domain or a full path - we normalize it for you.

https://www.yourdomain.com/answerA specific answer page is ideal; we read its entity and answer structure.

Avoid this

Expecting a single 'AI score'We report direct check counts and evidence by group, never one opaque number.

Public methodology

Exactly how this result is produced

The initial HTML is fetched and checked for directly observable content, heading, citation and structured-data signals. The root robots.txt and llms.txt files are captured as separate evidence when available. Each check remains visible on its own; no composite score or third-party behavior is inferred.

We fetch the initial HTML and inspect observable retrieval, answer, attribution and topic-structure signals.
We capture root robots.txt and llms.txt as separate evidence when available.
We report direct passed and measured counts for each group, with no composite score and no claim that an answer engine cited the page.

Built on public standards

The international standards this check applies

HTML and schema.org define the markup inspected, RFC 9309 defines robots.txt syntax, and llms.txt is labelled as a proposal rather than a standard or permission system. These references explain page signals only; they do not authorize a claim about how an answer engine behaved.

schema.orgStructured data

Structured data vocabulary

Reads declared entity and content types as one observable page signal.

Read the specification

Proposalllms.txt

llms.txt proposal

Captures /llms.txt when present and labels it as a proposal, not a universal permission standard.

Read the specification

IETFRFC 9309

Robots Exclusion Protocol

Reads captured robots.txt directives without claiming how a third party acted on them.

Read the specification

We list a standard only where this tool genuinely reads or measures against it. Where a signal is outside a live check, the result says so instead of implying coverage.

Common questions

GEO Checker FAQ

What does the GEO Checker measure?

It reads page HTML plus /llms.txt and /robots.txt and reports observable checks across Retrieve, Quote, Cite and Graph. It shows passed, review and gap counts without inventing a visibility score.

What do Retrieve, Quote, Cite and Graph mean?

Retrieve is whether engines can access the page, Quote is whether text is cleanly extractable, Cite is whether authorship and sources support attribution, and Graph is whether structured entities and schema make the content machine-understandable.

Does the GEO Checker prove ChatGPT or Perplexity cite my page?

No. It measures readiness signals in your page and configuration, not live engine behaviour. It does not query the AI engines themselves, so it cannot prove any specific citation actually happened.

What is /llms.txt and why does the checker read it?

/llms.txt is a proposed file signalling how AI systems may use your content. The checker reads it, alongside /robots.txt, to assess whether your page invites and permits generative engines to retrieve and reuse it.

How do I improve the reported GEO and AEO checks?

Review each failed or warning check. Use accurate structured data, clear self-contained answers, crawlable HTML and discovery rules that match the access you intend to permit.

What is the difference between GEO and traditional SEO?

Traditional SEO optimises for ranking positions in search results. GEO, generative engine optimization, optimises for being retrieved, quoted and cited inside AI-generated answers. GEO emphasises extractable, attributable, entity-rich content over link position.

Why does AEO readiness matter for my content?

Answer engine optimization improves the odds your content is surfaced within AI answers where users increasingly stop before clicking. Readiness signals make your page easier to retrieve, quote and attribute, protecting visibility as search shifts.

Why is my page hard for AI engines to retrieve?

Common causes include content rendered only by scripts, blocked crawlers in /robots.txt, thin HTML, or missing structure. The checker flags weak signals so you can expose clean, permitted, machine-readable content instead.

Do passed readiness checks guarantee AI citations?

No. They only confirm that the measured page-level signals were present. A citation is a separate observed outcome, which this checker does not query or claim.

How is Quote readiness different from Cite readiness?

Quote readiness is whether an engine can cleanly extract self-contained text to reuse. Cite readiness is whether clear authorship, dates and sources let an engine attribute that text to you. Both together support trustworthy AI answers.

More free checks